Science fact or urban legend?   If you get lost, you will walk in circles, just like you have seen in too many movies and TV shows to count.

Fact, say researchers from the Multisensory Perception and Action Group at the Max Planck Institute for Biological Cybernetics in Tübingen writing in Current Biology, though there was no scientific evidence to back that up until now.

And those circular paths are rarely systematic, the researchers say. The same person may sometimes veer to the left, then again to the right, before ending up back where they started from, which rules out one potential explanation for the phenomenon, namely that circle-walking stems from some systematic bias to turn in one direction, such as differences in leg length or strength.

It seems that the circles rather emerge naturally through "random drift" in where an individual thinks straight ahead to be.

Marc Ernst, Group Leader at the MPI for Biological Cybernetics, said, "The results from these experiments show that even though people may be convinced that they are walking in a straight line, their perception is not always reliable. Additional, more cognitive, strategies are necessary to really walk in a straight line. People need to use reliable cues for walking direction in their environment, for example a tower or mountain in the distance, or the position of the sun."

"The stories about people who end up walking in circles when lost are actually true," said Jan Souman of the Max Planck Institute for Biological Cybernetics in Germany. "People cannot walk in a straight line if they do not have absolute references, such as a tower or a mountain in the distance or the sun or moon, and often end up walking in circles." 

The researchers tested the idea in both forest and desert environments. Participants were instructed to walk as straight as they could in one direction, and their trajectory was recorded via GPS. Six people walked for several hours in a large, flat forest—four on a cloudy day with the sun hidden. Those four all walked in circles, with three of them repeatedly crossing their own paths without noticing it. In contrast, when the sun was out, two other participants followed an almost perfectly straight course, except during the first 15 minutes, when the sun was still hidden behind some clouds.

Three other participants walked for several hours in the Sahara desert, in southern Tunisia. Two of them, who walked during the heat of the day, veered from the course they were instructed to follow but did not walk in circles. The third walked at night, at first by the light of a full moon. Only after the moon disappeared behind the clouds did he make several sharp turns, bringing him back in the direction he started from.

In other tests, blindfolded people walked in surprisingly small circles, though rarely showing a tendency to travel in any particular direction. That result led the researchers to suggest that the inability to stick to a straight course results from accumulating "noise" in the sensorimotor system. Without an external directional reference to recalibrate the subjective sense of straight ahead, that "noise" may cause people to walk in circles, the researchers said.

Souman's group plans to study this tendency under more controlled conditions by asking subjects to walk through a virtual-reality forest on a special treadmill they have built, which allows a person to travel in any direction they choose. These tests will make it possible for the researchers to isolate the various factors that might play a role, such as the availability of the sun or other landmarks, and to study their contributions to walking straight—or in circles.